Project Summary/Abstract We are in the midst of an unprecedented epidemic of type 2 diabetes (T2DM). The proportion of the population with T2DM has doubled in just one generation. Although the epidemic is primarily driven by obesity, only ~20% of obese individuals develop T2DM. Genetic factors play a major role in determining who among obese individuals will progress to develop T2DM. The heritability of T2DM has been estimated to be approximately 0.5. Genetic studies in humans have identified many loci that contribute to diabetes susceptibility. Nearly all of these loci are involved in ?-cell biology. However, these loci only account for a small part of the high heritability of T2DM. A guiding premise of this grant proposal is that an important source of the missing heritability of human diabetes genetics is the inabiliy to carry out in-depth mechanistic phenotyping of beta-cell function. Here, we propose to carry out deep phenotyping of pancreatic islets to identify genes and pathways that confer susceptibility to T2DM. The project involves a hybrid strategy, employing two complementary mouse genetic cohorts: 1) a population of outbred mice and 2) a panel of recombinant inbred (RI) strains, both derived from the same founder population. This population, The Collaborative Cross, was derived from eight founder strains that together capture a major part of the genetic variability available in inbred mouse strains. Our preliminary studies of the founder strains revealed a high degree of phenotypic diversity in traits related to diabetes susceptibility. These results predict that the traits we propose to study will have strong heritability. Our team consist of a laboratory with extensive experience in islet biology and diabetes genetics together with four statistical geneticists who have developed many of the widely used methods for QTL mapping and causal network construction. We will carry out studies of insulin secretion, beta-cell proliferation and oxidative metabolism. In addition, we will conduct transcriptomic, proteomic, and phosphoproteomic studies on isolated pancreatic islets. In addition to classical association mapping, we will use genetic association data to construct causal networks linking gene loci with intermediate traits and disease phenotypes. Based on our prior experience with successful positional cloning projects, we predict that the loci we identify will be highly relevan to diabetes susceptibility in humans.